gpt-oss-20b

apache-2.020B

`gpt-oss-20b` is our medium-sized open-weight model for low latency, local, or specialized use-cases (21B parameters with 3.6B active parameters).

Intelligence
Reasoning
Speed
Context
131K
Max output
131K
Input price
$0.07/1M tokens
Output price
$0.3/1M tokens

Capabilities

Reasoning & Interaction
Reasoning
Tool Calling
Output Control
Structured Output
JSON Mode
Infrastructure
Streaming
Batch Processing

Details

Provider Vercel AI Gateway
Creator OpenAI
Familygpt-oss
Licenseapache-2.0
Statusactive
Typechat
Input modalitiestext
Output modalitiestext
Open weightYes
Parameters20B
Architecture—
Context window131K tokens
Max input—
Max output131K tokens
Reasoning tokensYes
Knowledge cutoff
Release date—
Sourceofficial

Tools

Function Callingfunction_calling
Call external functions and APIs

Pricing

Input
$0.07
Output
$0.3
Cache write
—
Cache read
—
Batch in
—
Batch out
—

Family Comparison: gpt-oss

ModelContextMax outPricing
gpt-oss-20b131K131K$0.07$0.3
gpt-oss-safeguard-20b131K—$0.07$0.3
gpt-oss-safeguard-120b128K—$0.15$0.6
gpt-oss-120b131K131K$0.35$0.75

API

GET/v1/models/vercel/openai/gpt-oss-20b

gpt-oss-20b

`gpt-oss-20b` is our medium-sized open-weight model for low latency, local, or specialized use-cases (21B parameters with 3.6B active parameters).

Changes · 3 entries
openai/gpt-oss-20bupdateb301ebeMar 23, 2026, 11:47 PM
modalitiesinput["text"]output["text"]
openai/gpt-oss-20bupdate2b100acMar 23, 2026, 05:30 AM
model_typechat
parameters20
tools["function_calling"]
openai/gpt-oss-20bcreate7f5c66fMar 21, 2026, 05:16 AM